Rev Quant Finan Acc

https://doi.org/10.1007/s11156-017-0695-0

ORIGINAL RESEARCH

Managerial ability and firm risk‑taking behavior

Kenneth Yung1 · Chen Chen1

© Springer Science+Business Media, LLC, part of Springer Nature 2017

Abstract  In this study, we show that managerial heterogeneity plays an important role in
firm decisions. Our view is that in addition to the effects of previously examined determi-
nants, firm decisions are affected not just by the managers’ explicit mandate to maximize
firm value, but also by the ability of the manager in managing the firm. We find that high-
ability managers and low-ability managers have opposite effects on firm behavior and firm
value. High-ability managers are receptive to risk-taking whereas low-ability managers
refrain from risk-taking. High-ability managers cut capital expenditures but spend signifi-
cantly more on research and development projects; low-ability managers reduce both capi-
tal expenditures and research and development expenses significantly. High-ability manag-
ers are associated with higher levels of firm focus than low-ability managers. Managerial
ability is negatively associated with firm leverage. In addition, our results show that high-
ability managers are associated with increases in firm value whereas low-ability managers
are associated with decreases in firm value.

Keywords  Managerial ability · Managerial attributes · Risk-taking · Firm value ·

Managerial incentives · Corporate governance

JEL Classification  G31 · G32 · G34

1 Introduction

Traditional models assume firm managers behave rationally and follow the mandate of
firm value maximization; managerial heterogeneity is accordingly considered unlikely to

* Kenneth Yung
kyung@odu.edu
Chen Chen
cchen027@odu.edu
1
Department of Finance, Srome College of Business, Old Dominion University, Norfolk,
VA 23520, USA

13
 K. Yung, C. Chen

have a significant role in firm decisions (Berk and Stanton 2007; Bamber et al. 2010). Yet,
recent advances of the research of behavioral finance have amassed evidence that manage-
rial attributes are important determinants of firm behavior (Graham et  al. 2012; Kaplan
et al. 2012). Among these studies, it has been shown that managerial ability has significant
impact on corporate earnings quality (Demerjian et al. 2013; Choi et al. 2015); firm inno-
vative activity (Chen et al. 2015); bank liquidity creation (Andreou et al. 2016); and strate-
gic entry in a new market (Goldfarb and Xiao 2011).
The literature on managerial ability is relatively new and infrequent. In this study, we
extend the literature by investigating the effect of managerial ability on firm risk-taking
behavior. Corporate risk-taking is fundamental to firm survival. Researchers have long
argued that managers’ willingness to take risks in the pursuit of profitable opportunities is
a fundamental driving force of firm performance and growth (Bromiley 1991; John et al.
2008). There is also evidence that many managers believe that risk-taking is an essential
element of the managerial role (March and Shapira 1987). Surveys of business executives
conducted by earlier investigations found that managerial risk-taking propensities vary
across individuals, and the variation across managers appeared to be related to personal-
ity traits and experience (Maccrimmon and Wehrung 1986; Shapira 1986). Based on the
findings of Nuthall (2001, 2009) that personality traits, early life experience, education and
intelligence are determinants of managerial ability; one can posit that managerial risk-tak-
ing propensity is related to managerial ability. Koijen (2014) finds that there is substantial
heterogeneity in the ability of the managers in the investment industry. Laboratory experi-
ments also suggest that there is significant heterogeneity in the ability of decision makers
(Camerer 2003; Camerer et al. 2004; Costa-Gomes and Crawford 2006). Managerial abil-
ity is likely a valuable input to the firm. Consistent with this view, Murphy and Zabojnik
(2007) and Custódio et al. (2013) find that managerial ability is a sought after asset as firms
frequently offer lucrative packages to lure capable CEOs from other companies.
Chen et  al. (2015) find that managerial ability is positively related to firm innovative
activity. Admittedly, firm innovation is an important form of corporate risk-taking; but
with firm innovation measured by the number of patent grants or citations, it is likely that
firm innovation tends to concentrate in a few industries only. In fact, Fang et  al. (2014)
report that firm-year observations with zero patents represent 77% of their sample. Simi-
larly, Tian and Wang (2014) report that 73% of their sample observations have zero pat-
ents. According to the distribution statistics reported by Fang et al. (2014), innovation is
concentrated in a few industries such as manufacturing, healthcare, chemicals, and busi-
ness equipment. The literature has not yet examined the effect of managerial ability on low-
risk investment decisions such as capital expenditures (Bhagat and Welch 1995). Similar,
not much is known regarding the effect of managerial ability on research and development
that do not result in patent grants. The investigation of Chen et  al. (2015) therefore is a
limited study of the effect of managerial ability on firm risk-taking activity. In this study,
we expand the literature by examining firm risk-taking behavior in general. We measure
firm risk-taking by a number of proxies other than innovation and investigate the effect of
managerial ability on these general risk-taking activities. Despite the findings of Nuthall
(2001, 2009) suggest that managerial ability is positively related to education background
and intelligence, some behavioral studies have provided evidence that educated persons
and individuals of higher levels of intelligence are more risk cautious in making decisions
(Sjöberg and Drottz-Sjöberg 1991; Boholm 1998; Culver et al. 2001). In addition, Wang
et al. (2013) find a significant negative relation between the education background of busi-
ness executives and corporate risk-taking in China. Thus, the effect of managerial ability
on firm risk-taking is less than straightforward.

13
Managerial ability and firm risk‑taking behavior

In this study, we measure firm risk-taking behavior by the standard deviation of return
on assets (σROA), the standard deviation of return on equity (σROE), capital expenditures to
total assets (CAPEX_TA), research and development expenses to total assets (R&D_TA),
acquisitions value to total assets (ACQ_TA), sales-based Herfindahl index, and book lever-
age, respectively. We obtain several important results in this study. First, our results show
that managerial ability is associated with higher levels of risk-taking activity except firm
leverage. Second, the effect of managerial ability on firm risk-taking is non-linear. We find
that managers of high-ability are in general receptive to risk-taking whereas managers of
low-ability are unwilling to take risks. Third, our results show that high-ability manag-
ers shift resources from low risk capital expenditures to high risk research and develop-
ment projects whereas low-ability managers significantly reduce capital expenditures and
research and development expenses. Fourth, managers of high-ability are associated with
higher levels of firm focus whereas managers of low-ability are associated with lower
levels of firm focus. Fifth, firms with high-ability managers have lower leverage ratios
whereas firms with low-ability managers have higher leverage ratios. Sixth, we find that
high-ability managers are associated with higher levels of Tobin’s Q and low-ability man-
agers are associated with lower levels of Tobin’s Q. The economic impact of managerial
ability on firm value is significant.
We implement additional tests to ensure our results are robust. First, we address the
effect of managerial incentives as it has been argued that CEO compensation can be struc-
tured to influence the risk-taking behavior of managers (Coles et  al. 2006). Second, we
further control for the effect of corporate governance given the fact that firm risk-taking
behavior is significantly related to governance practices (John et al. 2008). The additional
tests show that our results remain robust.
This study is related to a growing literature on the role of managers in firm performance
(Baker and Wurgler 2012). This literature has examined how corporate achievements
relate to management practices and characteristics (Bloom and Van Reenen 2009), inter-
personal skills (Kaplan et  al. 2012), overconfidence (Malmendier and Tate 2005, 2008),
and manager education and experience (Bertrand and Schoar 2003). However, the focus on
managerial attributes overlooks the effect of managerial ability in decision making. In this
study, we make three contributions to the nascent literature on managerial ability. First, our
results on the effects of managerial ability on capital expenditures, R&D expenses, firm
acquisitions, firm focus, and leverage are new additions to the literature. Existing studies
have only examined the effects of managerial ability on firm innovation and bank liquidity
creation. Second, in contrast to earlier studies that show that high-ability managers increase
risk-taking more whereas low-ability managers increase risk-taking less, our results show
that high-ability managers and low-ability managers have opposite effects on firm risk-
taking and firm value. We document that low-ability managers are in fact associated with
decreases in firm risk-taking activity and firm value. Third, our finding of a negative asso-
ciation between high-ability managers and firm leverage adds to the literature that mana-
gerial attributes have important impacts on financing decisions (Chemmanur et al. 2009;
Bhagat et al. 2012). Chemmanur et al. (2009) argue that managers of higher reputation can
overcome problems of asymmetric information more effectively and thus prefer to use less
debt because they have better access to equity markets. Our results, however, suggest that
the lower firm leverage associated with high-ability managers is consistent with the sug-
gestion of behavioral researchers that capable individuals dislike third party monitoring.
We examine the relationship between managerial ability and the risk choices in corporate
investment in this study. Our view is that in addition to the effects of previously examined

13
 K. Yung, C. Chen

determinants, the risk choices are affected not just by the managers’ explicit mandate to
maximize firm value, but also by the ability of the manager in managing the firm.
The reminder of the paper is structured as follows. We review the related literature and
present our empirical hypotheses in Sect. 2. Section 3 describes key variable construction
and our sample. Section 4 presents our main results and provides some interpretation. Sec-
tion 5 implements some robustness tests and examines the effect of managerial ability on
firm value. Section 6 provides a summary.

2 Related literature and hypothesis development

The importance of managers for the outcomes achieved by the firm is emphasized by
researchers of organization behavior. For example, the upper echelons theory argues that
the complexity of actual decision-making situations demands an idiosyncratic importance
of the top managers (Hambrick and Mason 1984; Hambrick 2007). Bertrand and Schoar
(2003) report that managers influence their organization’s behavior over and above time-
and firm-specific characteristics. Complementing the findings reported in the literature
of organization behavior, recent studies conducted by financial economists find evidence
that managerial characteristics have significant impact on firm behavior. For example, Bol-
ton et al. find that CEOs with strong managerial resoluteness perform better in situations
requiring greater coordination. Malmendier and Tate (2005, 2008) find that overconfident
CEOs are more likely to engage in value-destroying mergers and acquisitions. Yim (2013)
finds that CEO age has a negative effect on firm investment activity. Bargeron et al. (2010)
suggest that CEO risk preference causes significant declines in firm investment activity
after the implementation of the Sarbanes–Oxley Act of 2002. Chemmanur et  al. (2009)
find that managerial quality and CEO reputation significantly impacts the firm investment.
The significance of managerial ability in decision making has been documented in
behavioral studies. Numerous laboratory experiments show that some people are better
than others in decision making as they have the ability to execute better strategies; this
heterogeneity does not appear to be random (Camerer 2003; Camerer et al. 2004; Costa-
Gomes and Crawford 2006). Financial economists attest to the importance of managerial
ability in some recent studies. It has been shown that managerial ability has significant
impact on corporate earnings quality (Demerjian et al. 2013; Choi et al. 2015); firm inno-
vative activity (Chen et al. 2015); bank liquidity creation (Andreou et al. 2016); and strate-
gic entry in a new market (Goldfarb and Xiao 2011).
Existing studies justify a positive association between managerial ability and cor-
porate risk-taking. For example, it has been shown that skillful investment managers are
also active managers who can make superior risky investment decisions in the investment
industry (Cremers and Petajisto 2009). Chemmanur et al. (2009) find that managers with
better reputation and managers of higher quality are capable of selecting better projects;
the authors further argue that firms with better managers are likely to have a larger equilib-
rium scale of investment. Accordingly, it is likely that capable managers are more willing
to assume risks in corporate decisions. Consistent with this view, Chen et al. (2015) and
Andreou et  al. (2016) find that capable managers are more inclined to pursue corporate
risk-taking activity.
The literature also offers justifications for a negative association between manage-
rial ability and risk-taking. Nuthall (2001, 2009) suggests that managerial ability is posi-
tively associated with education background and the level of intelligence; however, some

13
Managerial ability and firm risk‑taking behavior

behavioral studies suggest that individuals with significant education and/or high levels of
intelligence are more risk cautious and conservative in assuming risk (Sjöberg and Drottz-
Sjöberg 1991; Boholm 1998; Culver et al. 2001). Thus, capable managers, likely educated
and intelligent, may be less inclined to take risks in firm decisions. Consistent with this
view, Wang et al. (2013) find a significant negative relation between the education back-
ground of business executives and corporate risk-taking in China. Mishra (2014) argues
that when a manager’s skills are highly firm-specific, concerns regarding the protection of
his human capital may cause the manager to resist risk-taking. A similar opinion is made
by Amihud and Lev (1981). In addition, it has been found that some managers may avoid
risk-taking because they prefer a quiet-life (Bertrand and Mullainathan 2003) or want to
play it safe (Gormley and Matsa 2016).
The mixed theoretical possibilities regarding the relation between managerial ability
and firm risk-taking motivate our empirical investigation. As managerial ability is likely
an important determinant of corporate decisions, its effect on firm value cannot be ignored.
Thus, we also investigate the relation between managerial ability and firm value.
Earlier studies report a positive relation between management quality and firm perfor-
mance (Chemmanur et al. 2009). In addition, Demerjian et al. (2013) show that managerial
ability enhances firm operating performance. Thus, existing results imply that managerial
ability has a positive impact on firm value. On the other hand, Mishra (2014) argues that
capable managers, given their greater mobility in the job market, are relatively less con-
cerned about firm-specific risk and are more likely to engage in value-destroying activity
because the personal objectives of these managers may be different from the interests of
shareholders. Accordingly, the argument of Mishra (2014) implies that managerial abil-
ity could have a negative effect on firm value. Moreover, if educated and intelligent indi-
viduals are more risk cautious as suggested by some behavioral studies (Halek and Eisen-
hauer 2001; Culver et  al. 2001), the educated and intelligent capable managers may be
less inclined to pursue value-enhancing risk-taking activity. In sum, opposite views can
be found in the literature regarding the relation between managerial ability and firm value.
Based on the discussion presented above, we develop the following hypotheses:

Hypothesis 1  High-ability (low-ability) managers are associated with increases

(decreases) in firm risk-taking activity.

Hypothesis 2  High-ability (low-ability) managers are associated with increases

(decreases) in firm value.

3 Key variables and the sample

3.1 Measuring managerial ability

To infer managerial ability, prior studies rely on proxies such as firm size, past abnor-
mal performance, compensation, tenure, media coverage, education, or manager fixed
effects. These measures, however, are noisy in general. For example, media coverage is
more common for large firms; abnormal stock returns can be attributed to market factors
other than managerial ability. Although manager fixed effects are more directly attribut-
able to management, they can be applied only to a relatively small sample of firms and do
not offer a stand-alone measure of ability (Demerjian et al. 2012). We follow the two-step

13
 K. Yung, C. Chen

methodology of Demerjian et  al. (2012, 2013) in measuring managerial ability. The first
step is to use data envelopment analysis (DEA) to create an initial measure of the relative
efficiency of the firm within its industry (Charnes et al. 1978). DEA is a linear program-
ming methodology that measures the relative efficiency of decision-making units (firms)
by evaluating inputs (labor, capital, etc.) relative to outputs (revenue, income, etc.). Effi-
cient firms are those that generate more revenues from a given set of inputs. The following
optimization is applied to estimate firm efficiency:
Sales
maxV θ =
v1 CoGS + v2 SG&A + v3 PPE + v4 OpsLease + v5 R&D + v6 Goodwill + v7 OtherIntan

where CoGS is cost of goods sold; SG&A is selling and administrative expenses; PPE is
net PP&E; OpsLease is net operating leases; R&D is net research and development; Good-
will is purchased goodwill; and OtherIntan is other intangible assets. The firm efficiency
measure, however, is affected by both firm-specific factors and management characteristics.
The second step is to remove firm-specific characteristics from the DEA generated firm
efficiency measure. This is done by removing the effects of firm size, market share, free
cash flow, firm age, multi-segment and international operations challenges by performing
the following regression:
Firm Efficiencyi = α0 + α1 Ln(Total Assets)i + α2 Market Sharei
+ α3 Free Cash Flow Indicatori
+ α4 Ln(Firm Age)i + α5 Business Segment Concentrationi (1)
+ α6 Foreign Currency Indicatori + α7 Year Indicatori + εi

According to Demerjian et al. (2012), the error term of the regression measures mana-
gerial ability.

3.2 Measuring firm risk‑taking behavior

To measure corporate risk-taking behavior, we rely on seven widely used measures in

the literature: standard deviation of return on assets (σROA), standard deviation of return
on equity (σROE), capital expenditures to total assets (CAPEX_TA), research and devel-
opment expenses to total assets (R&D_TA), acquisitions value to total assets (ACQ_TA),
sales-based Herfindahl Index, and book leverage. σROA (σROE) is computed over a 5-year
rolling window. σROE has been used in previous studies as an indicator of firm riskiness
(Faccio et  al. 2011); it reflects both the riskiness of a firm’s projects and the additional
risk associated with the use of leverage in the capital structure. While CAPEX_TA is com-
monly used in the literature as a proxy for risk-taking, Bhagat and Welch (1995) suggest
that capital expenditures represent low-risk activity. Following Berger and Ofek (1995), we
compute sales-based Herfindahl Index based on the Fama–French 48 industrial classifica-
tions, 2-digit SIC code, and 4-digit SIC code, respectively. Herfindahl Index is an indica-
tion of industry concentration. A high Herfindahl Index implies a high level of firm focus
and fewer business segments (Berger and Ofek 1995). Firms that are focused are likely to
be affected significantly by external shocks due to non-diversified cash flows. ACQ_TA
(Acquisition expenditures to total assets) is a risk-taking measure because acquisitions fre-
quently involve significant amounts of resources and have a high rate of failure (Porter
1987).

13
Managerial ability and firm risk‑taking behavior

3.3 Sample, data, and descriptive statistics

Our sample consists of all the firms on Compustat between 1980 and 2014. We exclude
depository receipts and firms in utility and financial industries. Following Bates et  al.
(2009), research and development expenses (R&D) is assigned a value of zero if the num-
ber is missing. We also exclude observations that have negative total assets or stockholders’
equity.
We select our control variables following previous literature. Specifically, we use firm
level controls, CEO characteristics controls, and corporate governance controls. For firm
level control variables, we use return on assets (ROA), sales growth, firm size, firm age,
market-to-book ratio, surplus cash ratio and dividend cut. To calculate the market value of
a firm, we use the sum of market value of equity and book value of labilities. In robustness
checks, we control for managerial incentives. Specifically, we take into consideration the
influence of the CEO delta and CEO vega. CEO delta and vega are used to capture CEOs’
risk attitudes. As in previous literature, CEO delta is defined as the change in dollar value
of CEOs’ wealth for one percentage point change in stock price; CEO vega is the change
in the dollar value of CEOs’ wealth for 0.01 change in the annualized stock volatility. We
calculate CEO delta and CEO vega following Guay (1999) and Core and Guay (2002). The
logarithmic values of CEO delta and CEO vega are used in our regression analysis. For
controlling the impact of corporate governance, we use board size as our proxy. We count
the number of board directors per year in ExecuComp as the board size.
Values of dependent and independent variables are collected from several sources. Spe-
cifically, we collect firm-specific and industry segment data from Compustat. CEO com-
pensation, characteristics of the CEO, and the number of board members are collected
from ExecuComp. Because the data of ExecuComp start after 1992, the number of obser-
vations may be smaller in some regressions. We use stock return data from the Center for
Research in Security Prices (CRSP) to estimate expected stock return volatility and then
use the Black–Scholes option pricing model to calculate CEO delta and CEO vega. We
winsorize the variables at the 1st and 99th percentiles. Our final sample has 130,317 firm-
year observations. In Table 1, we provide a list of the variables used in this study and their
definitions.
Table  2 presents descriptive statistics of the sample. Included are the mean, median,
standard deviation, and 25th and 75th percentiles. The mean(median) managerial ability
score is 0.00 (−  0.01). Demerjian et  al. (2012) report a mean (median) managerial abil-
ity score of − 0.004 (− 0.013) for US firms between 1992 and 2009. In Table 2, the mean
(median) CEO cash compensation is 1.04 (0.80) million, the mean (median) CEO age is
55.4 (55) years old, and the mean(median) CEO tenure is 7.94 (5.67) years. These numbers
are comparable to those reported in Demerjian et  al. (2012). Regarding the measures of
firm risk-taking, the mean (median) σROA is 0.07 (0.05), the mean (median) σROE is 0.18
(0.10) and the mean (median) Acquisition Ratio is 0.02 (0.00). CAPEX_TA has a mean of
0.04 and a median of 0.02; R&D_TA has a mean of 0.04 and a median of 0.00; Herfindahl
Index has a mean of 0.89 and a median of 1.00; Book leverage has a mean of 0.22 and a
median of 0.20. The statistics are consistent with prior studies (Coles et al. 2006).

13
 K. Yung, C. Chen

Table 1  Variables definition
Variable Definition

σ(ROA) Standard deviation of returns on assets over time-period (t, t + 4)

σ(ROE) Standard deviation of returns on equity over time-period (t, t + 4)
CAPEX_TA Capital expenditures to total assets
R&D_TA Research and development expenses to total assets
Acquisition_TA Acquisitions to total assets
Herfindahl index-2Digital SIC Herfindahl Index = (Sum of squared segment sales)/(squared firm sales): (The
segment is based on 2 digital SIC code)
Herfindahl index-4 Digital SIC Herfindahl Index = (Sum of squared segment sales)/(squared firm sales): (The
segment is based on 4 digital SIC code)
Herfindahl index-ff48 Herfindahl Index = (Sum of squared segment sales)/(squared firm sales): (The
segment is based on Fama–French 48 Industries category)
Book Leverage (Long-Term Debt + Debt in Current Liabilities)/Total Assets
Managerial Ability Managerial ability score (Demerjian et al. 2012, 2013)
CEO Cash Compensation CEO total current compensation (salary + bonus)
CEO Delta The CEO Delta is defined as the change in dollar value of CEOs’ wealth for one
percentage point change in stock price
CEO Vega The CEO Vega is the change in the dollar value of CEOs’ wealth for 0.01
change in the annualized stock volatility.
CEO Equity Holding Value of CEOs’ Equity Holding: CEO Total Stock Value + CEO Total Option
Value
CEO tenure Dummy variable that equals one if the CEO was replaced, and zero otherwise
ROA Returns on assets: income before extraordinary items/total assets
ROE Returns on equity: income before extraordinary items/total stockholders’ equity
Firm Size Logarithmic value of total sales
Firm Age Logarithmic value of (1 + number of years since public)
Market-to-book (Fiscal Annual Closed Price * Common Shares Outstanding + Total Assets -
Total Common Equity)/Total Assets
Surplus Cash/Total Assets (Operating Activities Net Cash Flow − Depreciation and Amortization + R&D
Expense)/Total Assets
Sales Growth ln
Salest
Salest−1
Stock return Stock’s annual return
Net fixed assets Net property, plant, and equipment scaled by total assets (Net PPE/TA)
Board Size The number of board directors reported in ExecuComp
Dividend Cut Dummy variable that equals one if there is a reduction in annual dividend, and
zero otherwise
Z-Score (1.2*(Total Current Assets − Total Current Liabilities) + 1.4* Retained Earn-
ings + 3.3*(Net Income (Loss) + Total Interest and Related Expense + Total
Income Taxes) + Sale)/Total Assets +0.6* Fiscal Annual Closed Price *
Common Shares Outstanding/Total Liabilities
EBIT_TA Earnings before interest and taxes to total assets

4 Main results

4.1 Managerial ability and firm risk‑taking behavior

To analyze the impact of managerial ability on firm risk-taking behavior, we aug-

ment standard firm risk-taking regression models by adding managerial ability as the

13
Managerial ability and firm risk‑taking behavior

Table 2  Descriptive statistics of the sample

Variables N Mean SD 25th 50th 75th

σ(ROA) 74,054 0.07 0.07 0.02 0.05 0.10

σ(ROE) 74,047 0.18 0.20 0.05 0.10 0.22
CAPEX_TA 131,037 0.04 0.06 0.00 0.02 0.06
R&D_TA 131,037 0.04 0.08 0.00 0.00 0.04
Acquisition_TA 125,999 0.02 0.06 0.00 0.00 0.00
Herfindahl index-2Digital SIC 125,054 0.91 0.18 1.00 1.00 1.00
Herfindahl index-4 Digital SIC 125,054 0.84 0.24 0.65 1.00 1.00
Herfindahl index-ff48 125,054 0.91 0.18 1.00 1.00 1.00
Book Leverage 130,835 0.22 0.20 0.04 0.20 0.35
Managerial Ability(t − 1) 131,037 0.00 0.10 − 0.06 − 0.01 0.04
CEO Cash Compensation 28,498 1044.00 872.30 521.36 802.00 1210.54
CEO Delta(t − 1) 27,986 634.68 1483.28 65.80 183.45 527.77
CEO Vega(t − 1) 27,986 111.19 202.69 7.11 35.90 114.82
CEO Equity Holding 28,498 54,037.11 146,101.47 3933.80 12,623.71 37,679.80
CEO Age 28,313 55.37 7.66 50.00 55.00 60.00
CEO tenure 27,965 7.94 7.61 2.67 5.67 10.67
ROA 131,037 − 0.02 0.21 − 0.03 0.03 0.07
ROE 131,037 − 0.14 0.81 − 0.08 0.07 0.15
Ln(Sale) 131,037 4.88 2.21 3.26 4.83 6.43
Firm Age 131,037 15.09 10.77 6.00 12.00 21.00
Market Value/Book Value 131,037 1.78 1.47 0.98 1.31 1.98
Surplus Cash/Total Assets 103,039 0.03 0.16 − 0.02 0.04 0.11
Sales Growth 131,024 0.10 0.33 − 0.03 0.08 0.21
Stock returns 114,052 − 0.01 0.57 − 0.29 0.04 0.32
Board Size 11,452 8.00 2.45 6.00 8.00 10.00

independent variable of interest. Specifically, our model has the following specification:
Risk − taking activityit = 𝛽0i + 𝛽1i Managerial Abilityit−1 + 𝛽2i Xit
(2)
+ Industry dummies + Year dummies + εit

where risk-taking activity is measured by the standard deviation of ROA(σROA), the stand-
ard deviation of ROE(σROE), capital expenditures divided by total assets (CAPEX_TA),
research and development expenses divided by total assets(R&D_TA), acquisition values
to total assets (ACQ_TA), Herfindahl Index, and book leverage, respectively.
Following existing studies, the control variables, ­Xit, include (1) firm age, measured by
Ln(1 + firm age); (2) stock return, measured by the prior year buy-and-hold stock return;
(3) firm performance, measured by return on assets; (4) investment opportunities, meas-
ured by Ln(M/B ratio); (5) book leverage; (6) firm size, a proxy for economies of scale, is
measured by log(sales); (7) sales growth; (8) surplus cash, defined as the amount of cash
available to finance new projects, scaled by total assets (see Coles et  al. 2006); and (9)
dividend cut, a (0, 1) dummy variable that has a value of one if there was a decrease in the
annual dividend, and zero otherwise. The lagged value of managerial ability is used in the
estimation to control for potential endogeneity problems (Coles et al. 2006). We also con-
trol for industry fixed effects and year fixed effects in the model.

13
 K. Yung, C. Chen

In Table 3, regression results using Eq. (2) are reported. In columns (1) to (6) where the
dependent variable in each column is a measure of investment behavior, the coefficient on
managerial ability is positive and significant at the one percent level. In column (7) where
the dependent variable is book leverage, the coefficient on managerial ability is negative
and significant at the one percent level. In short, managerial ability is positively associated
with firm risk-taking behavior but negatively associated with firm leverage. The results in
Table 3 suggest that the economic impact of managerial ability on firm risk-taking is con-
siderable. For example, in column (1), the coefficient on managerial ability is 0.0268 with
a t-value of 9.25. Evaluating σROA at it mean, a one standard deviation increase in mana-
gerial ability is associated with a 3.83% increase in σROA. Similarly, evaluating σROE at it
mean, a one standard deviation increase in managerial ability is associated with a 4.29%
increase in σROE. The positive relation between managerial ability and the two risk meas-
ures (σROA and σROE) is consistent with the finding of Andreou et al. (2016) that managerial
ability increases the risk-taking activity of financial institutions. In addition, columns (3)
and (4) show that a one standard deviation increase in managerial ability is associated with
a small increase of 1.83% in CAPEX_TA but a large increase of 11.63% in R&D_TA. Bha-
gat and Welch (1995) argue that capital expenditures represent low risk investment activity.
The results in columns (3) and (4) show that managerial ability is more associated with
high risk research and development than low risk capital expenditures. Consistent with
our results, Chen et al. (2015) find that firms with capable managers are associated with
higher levels of innovation, which are in general associated with research and development
expenses. Our observation of a positive association between managerial ability and firm
investment activity does not support the implications of some behavioral studies that man-
agers, because of their higher levels of education and intelligence, may be more risk cau-
tious. In column (5), the regression result shows that a one standard deviation increase in
managerial ability is positively associated with a 4.4% increase in firm acquisitions. Some
researchers have documented that managerial characteristics are related to firm acquisi-
tion activity (Bertrand and Schoar 2003; Malmendier and Tate 2005; Jenter and Lewellen
2015). Our results add to this line of research by showing that managerial ability has a
positive effect on firm acquisitions. In column (6), the dependent variable is Herfindhal
Index. An increase in the value of Herfindhal Index implies a larger market share and a
higher market power; it happens when a firm increases its focus in fewer market segments.
In column (6), the coefficient on managerial ability is positive and significant at the one
percent level. The result suggests that managers with high-ability are associated with firms
that have revenues from fewer market segments; that is, managerial ability increases firm
focus. Earlier studies on firm efficiency in general conclude that improved firm focus has
positive effects on firm performance and firm value (Comment and Jarrell 1995). The liter-
ature, however, has not examined the relation between managerial characteristics and firm
focus. To the best of our knowledge, our result in column (6) is likely the first to report
a relation between managerial ability and firm focus. In sum, the results in Tables 3 and
4 suggest that firms with managerial ability is positively associated with higher levels of
risk-taking activity in general. It is interesting to note that in column (7), the result shows
that a one standard deviation increase in managerial ability is associated with a 4.59%
decrease in book leverage. Consistent with our finding, Bhagat et al. (2012) develop a the-
oretical construct of managerial ability and predict that managerial ability is associated
with a decline in firm leverage. According to Bhagat et al. (2012), capable managers prefer
less debt because firm leverage increases the expected bankruptcy costs for the firm and
personal bankruptcy costs for the manager, which negatively affect his continuation value.
Some behavioral studies suggest that capable individuals are less receptive of interferences

13
 Table 3  Regressions of managerial ability effect on firm risk-taking behavior
(1) (2) (3) (4) (5) (6) (7)
σ(ROA) σ(ROE) CAPEX_TA R&D_TA Acquisition_TA Herfindahl index-ff48 Book Leverage

Managerial Ability 0.0268*** 0.0772*** 0.0073*** 0.0465*** 0.0159*** 0.0389*** − 0.1010***

(9.25) (9.03) (3.52) (21.48) (7.80) (7.90) (− 20.19)
Book Leverage − 0.0281*** 0.1263*** 0.0003 − 0.0358*** 0.0436*** − 0.0234***
(− 16.76) (21.55) (0.26) (− 32.57) (32.55) (− 7.46)
Stock return − 0.0075*** − 0.0180*** − 0.0070*** − 0.0003 0.0030*** − 0.0046***
(− 11.30) (− 8.91) (− 16.82) (− 0.56) (7.48) (− 4.55)
ROA − 0.2175*** − 0.5209*** 0.0007 − 0.2715*** − 0.0063*** 0.0051* − 0.0367***
(− 72.15) (− 61.69) (0.49) (− 90.46) (− 4.39) (1.73) (− 9.80)
Ln(Sale) − 0.0058*** − 0.0099*** − 0.0006*** − 0.0028*** 0.0022*** − 0.0136*** 0.0118***
Managerial ability and firm risk‑taking behavior

(− 35.80) (− 19.95) (− 5.32) (− 24.96) (18.64) (− 36.42) (42.86)

Ln (Firm Age) − 0.0112*** − 0.0290*** − 0.0073*** − 0.0037*** − 0.0036*** − 0.0472*** − 0.0229***
(− 24.57) (− 21.02) (− 23.03) (− 13.13) (− 10.41) (− 51.70) (− 28.42)
Market-to-book Value/Book Value 0.0047*** 0.0118*** 0.0034*** 0.0059*** − 0.0028*** 0.0030*** 0.0127***
(17.58) (15.05) (21.04) (26.46) (− 20.22) (8.92) (25.14)
Surplus Cash/Total Assets 0.0099*** 0.0292*** 0.0235*** 0.2482*** 0.0157*** 0.0621***
(3.02) (2.95) (13.60) (76.03) (8.82) (16.18)
Sales Growth 0.0040*** 0.0041 0.0188*** 0.0059*** 0.0359*** − 0.0074***
(3.38) (1.16) (24.56) (6.42) (39.98) (− 4.59)
Dividend Cut 0.0005 − 0.0041
(0.61) (− 1.52)
R&D_TA − 0.4619***
(− 55.59)
Z-Score − 0.0129***
(− 93.55)
Net PPE/TA 0.1554***

13
(48.29)

Table 3  (continued)
(1) (2) (3) (4) (5) (6) (7)
σ(ROA) σ(ROE) CAPEX_TA R&D_TA Acquisition_TA Herfindahl index-ff48 Book Leverage

13
Year Fixed Effect Yes Yes Yes Yes Yes Yes Yes
FF48 Fixed Effect Yes Yes Yes Yes Yes Yes Yes
Observations 51,835 51,923 90,497 90,497 87,165 85,560 111,681
Adj R-Square 33.75% 24.09% 12.94% 60.03% 9.33% 14.97% 30.83%

***, **, and * denote the significance levels of 1, 5, and 10%, respectively
K. Yung, C. Chen
 Table 4  Compare the effects of high (top quartile) and low (both quartile) managerial ability on firm risk-taking behavior
(1) (2) (3) (4) (5) (6) (7)
σ(ROA) σ(ROE) CAPEX_TA R&D_TA Acquisition_TA Herfindahl index-ff48 Book Leverage

Hi_MA 0.0065*** 0.0160*** − 0.0018*** 0.0140*** 0.0002 0.0110*** − 0.0082***

Managerial ability and firm risk‑taking behavior

(9.51) (7.78) (− 3.87) (28.03) (0.36) (8.56) (− 7.03)

Low_MA − 0.0006 − 0.0066*** − 0.0047*** 0.0022*** − 0.0053*** 0.0058*** 0.0205***
(− 0.87) (− 3.10) (− 9.15) (4.71) (− 10.53) (4.17) (15.89)
Control Variables Yes Yes Yes Yes Yes Yes Yes
Year Fixed Effect Yes Yes Yes Yes Yes Yes Yes
FF48 Fixed Effect Yes Yes Yes Yes Yes Yes Yes
Observations 51,835 51,923 90,497 90,497 87,165 85,560 111,681
Adj R-Square 33.77% 24.10% 13.02% 60.17% 9.40% 14.98% 30.85%

***, **, and * denote the significance levels of 1, 5, and 10%, respectively

13
 K. Yung, C. Chen

(Bandura 1997). Accordingly, the negative relation between managerial ability and firm
leverage suggests that capable managers dislike the monitoring imposed by debtholders.
Although it is not the focus of this paper, Table 3 also provides information about other
conventional determinants of risk-taking activity. The regression coefficients on these
determinants have signs that are largely consistent with existing studies. For example, firm
age has a negative and significant coefficient as older firms are likely to have fewer growth
opportunities relative to younger firms. Similarly, the coefficient on sales growth and the
coefficient on market-to-book are positive and significant. The coefficient on surplus cash
is positive and significant. Consistent with Coles et  al. (2006), the coefficient on stock
return is negative and significant.
To garner additional insights into the effect of managerial ability on firm risk-taking, we
repeat the regressions using Eq. (2) to examine if the effect of managerial ability on firm
risk-taking is nonlinear. Specifically, we want to examine if high managerial ability (top
quartile) and low managerial ability (bottom quartile) have different impacts on firm risk-
taking behavior. We revise Eq. (2) to include indicator variables for firms that are in the top
and bottom quartiles of MA score.
Risk − taking activityit = 𝛽0i + 𝛽1i Hi_MAit−1 𝛽2i Low_MAit−1 + 𝛽3i Xit
(3)
+ Industry dummies + Year dummies + εit

For brevity sake, we report only the coefficients on high managerial ability (hi_MA)
and low managerial ability (low_MA) in Table 4. It can be seen in Table 4 that managers
of high managerial ability and managers of low managerial ability indeed affect firm risk-
taking behavior differently. In column (1) where the dependent variable is σROA, the coef-
ficient on hi_MA is positive and significant whereas the coefficient on low_MA is nega-
tive and insignificant. The existing literature has not provided an explanation of why less
capable managers do not engage firms in risk-taking activity, a plausible explanation is
that less capable managers may be more content with a quiet life (Bertrand and Mullaina-
than 2003) or prefer a playing it safe approach (Gormley and Matsa 2016) in risk-taking.
Consistent with the result in column (1), it can also be seen that in column (2) where the
dependent variable is σROE, the coefficient on hi_MA is positive and significant at the one
percent level whereas the coefficient on low_MA is negative and significant at the one per-
cent level. In columns (3) and (4), the different impacts of high managerial ability and low
managerial ability on capital expenditures and R&D expenses are considerable. In column
(3) where the dependent variable is CAPEX_TA, the coefficient on hi_MA is − 0.0018; in
column (4) where the dependent variable is R&D_TA, the coefficient on hi_MA is 0.014
with a t-value of 28.03. The results suggest that managers with high-ability (top quartile)
reduce low risk capital expenditures and shift firm resources to high risk research and
development. In contrast, the coefficient on low_MA in columns (3) and (4), suggests that
low-ability managers (bottom quartile) cut capital expenditures significantly more (260%
of the amount cut by hi_MA managers) but shift only a nominal amount (15.7% of the
amount added by hi_MA managers) of resources to R&D expenses. Recall earlier that we
obtained a positive coefficient on MA in column 3 (dependent variable is CAPEX_TA)
of Table 3, the result in column 3 of Table 4 implies that the positive coefficient on MA
in Table  3 is primarily driven by managers of median-ability (and confirmed by untabu-
lated results). The result in column (5) of Table 4 shows that hi_MA managers are asso-
ciated with higher levels of firm acquisitions. In column (6), it is shown that firms with
high-ability managers are associated with higher levels of firm focus. In sum, the results
in columns (1) to (6) of Table  4 overwhelmingly suggest that hi_MA managers increase

13
Managerial ability and firm risk‑taking behavior

investment related risk-taking activities whereas low_MA managers decrease investment

related risk-taking activities. Interestingly, in contrast to the result in Table 3, the result in
column (7) of Table 4 shows that hi_MA managers are associated with lower firm leverage
whereas low_MA managers are associated with higher firm leverage. This finding suggests
that low_MA managers do not seek to avoid the monitoring imposed by debtholders.

5 Additional tests

5.1 The effect of managerial incentives

It has been generally argued that CEOs prefer less risk than do more well diversified out-
side shareholders (Amihud and Lev 1981). In addition, some managers are averse to risk-
taking because they either prefer a quiet life or want to play it safe. There is also signifi-
cant evidence that managers are less willing to assume risk due to potential legal liabilities
(Bargeron et al. 2010). When CEOs are risk-averse, shareholders may suffer.
Researchers have suggested that stock and option based compensation incentives could
be used to encourage managerial risk-taking behavior (Guo et al. 2015). Prior research on
managerial incentives suggests that the two most important measures in capturing incen-
tives are the sensitivity of managerial wealth to stock price, that is, compensation delta,
and the sensitivity of managerial wealth to stock return volatility, that is, compensation
vega (Guay 1999; Core and Guay 2002).
Stock-based pay makes CEO wealth sensitive to movements in the underlying stock
price. Theoretically, higher delta helps align managerial behavior with the interests of
shareholders resulting in a convex payoff that encourages managers to work harder and
take on value enhancing risky projects (Jensen and Meckling 1976; Haugen and Senbet
1981). However, empirical findings suggest that the effect of delta on managerial risk-tak-
ing is ambiguous. On the one hand, an increase in the sensitivity of compensation to stock
price improves the alignment of interests between managers and shareholders, since both
benefit from an increase in the stock price. On the other hand, high deltas expose risk-
averse managers to greater risk, which prior research argues might deter managers from
increasing their preference for risk-taking (Amihud and Lev 1981; Lambert et  al. 1991;
Brick et al. 2012).
Recent empirical studies present evidence of a positive relationship between vega and
managerial risk-taking (Guay 1999; Coles et al. 2006; Low 2009). The underlying rationale
is that convex payoffs (e.g., those of stock options) provide managers strong incentives to
increase the return volatility of the underlying asset, which will induce managers to take
more risk to maximize their expected wealth. However, whether the convexity of higher
vega overcomes managerial risk aversion remains a theoretical and empirical question.
For instance, the effectiveness of option compensation depends on the managerial util-
ity function. Guay (1999) and Ross (2004) present models in which the concavity of the
manager’s utility function overcomes the convexity of the payoff causing managers to be
more risk averse. In addition, empirical studies frequently measure vega using the market
value (Black–Scholes value) of the stock option. This measure differs from the CEO’s cer-
tainty equivalent value (the value of riskless cash the CEO would trade for the risky asset).
So, managers unable to sell or hedge their options will value them differently than market
value based on their personal risk preferences (Lambert et al. 1991; Carpenter 2000; Ross
2004). Thus, the net effect of option compensation on firm risk is unclear.

13
 K. Yung, C. Chen

The empirical validity of the effects of vega and delta is beyond the scope of this study.
Nevertheless, we examine the robustness of our results on the relation between manage-
rial ability and risk-taking by controlling for the effect of managerial incentives. Follow-
ing the literature, we also add CEO characteristics such as cash compensation, age, ten-
ure, and turnover, to the list of control variables. The regression model has the following
specification:

Risk − taking activityit = 𝛽0i + 𝛽1i Managerial Abilityit−1

+ 𝛽2i CEO_Deltait−1 + 𝛽3i CEO_Vegait−1 (4)
+ 𝛽4i Xit + Industry dummies + Year dummies + εit

In Table 5, regression results using Eq. (4) are reported. The results are largely consist-
ent with the findings reported in Table 3. With the exception in column (3), Table 5 shows
that the coefficient on managerial ability is positive and significant at the one percent level
when the dependent variable is a measure of investment related risk-taking activity. That
is, managerial ability is positively associated with firm risk-taking even after controlling
for the effect of managerial incentives on CEO behavior. In column (3) where the depend-
ent variable is CAPEX_TA, the coefficient on MA is insignificant. The result, consistent
with our earlier findings in Table 4, suggests that capable managers do not add resources
to low risk capital expenditures; instead, as shown in column (4), capable managers prefer
to engage in high risk R&D activity. In column (7) where the dependent variable is book
leverage, the coefficient on managerial ability is negative and significant at the one percent
level.
Consistent with the existing literature, the results in Table 5 show that the effect of man-
agerial incentives on firm risk-taking is less than straight forward. Specifically, the sign of
the coefficient on delta in Table 5 does not remain consistent. The same is true of the sign
of the coefficient on vega. Nevertheless, the negative effect of delta and the positive effect
of delta on CAPEX are consistent with the findings of Coles et  al. (2006). Specifically,
Coles et al. (2006) find that greater sensitivity of CEO wealth to stock return volatility is
related to greater research and development expenditures and less capital expenditures.
We compare the effects of high managerial ability and low managerial ability and report
the results in Table  6. For brevity sake, we report only the coefficients on hi_MA and
low_MA in the table. As can be seen in Table  6, except for CAPEX_TA in column (3),
the coefficient on hi_MA is positive and significant in the first six columns. That is, hi_
MA managers are associated with increases in investment related risk-taking. In contrast,
the coefficient on low_MA in Table 6 is negative and significant in general; the results in
Table 6 show that low_MA managers are associated with decreases in investment related
risk-taking. More importantly, after controlling for the effect of managerial incentives, the
results in columns (3) and (4) show that hi_MA managers cut capital expenditures and
spend more on R&D expenses; low_MA managers cut both capital expenditures and R&D
expenses.

5.2 The effect of corporate governance

Agency theory predicts that the divergence of interests between shareholders and managers
has significant impacts on investment decisions. Yet the effect of agency conflicts on firm
risk-taking is ambiguous. On the one hand, some researchers argue that managers overin-
vest for their own benefits rather than the benefits of the firm’s shareholders (Jensen and

13
 Table 5  Regressions of managerial ability effect on firm risk-taking behavior while controlling for the effect of managerial incentives
(1) (2) (3) (4) (5) (6) (7)
σ(ROA) σ(ROE) CAPEX_TA R&D_TA Acquisition_TA Herfindahl index-ff48 Book Leverage

Managerial Ability 0.0209*** 0.0451*** 0.0001 0.0389*** 0.0194*** 0.0501*** − 0.1021***

(4.74) (3.36) (0.03) (15.46) (5.39) (5.75) (− 12.49)
Ln (CEO Delta) − 0.0001 − 0.0011 0.0016*** 0.0001 0.0007* − 0.0029*** − 0.0041***
(− 0.23) (− 0.86) (4.99) (0.24) (1.68) (− 2.83) (− 4.72)
Ln (CEO Vega) − 0.0005 0.0001 − 0.0011*** 0.0021*** − 0.0002 0.0028*** 0.0027***
(− 1.59) (0.06) (− 5.05) (14.52) (− 0.72) (3.69) (4.50)
Ln (CEO Age) − 0.0058 − 0.0154 − 0.0150*** − 0.0142*** 0.0055* − 0.0628*** − 0.0263***
Managerial ability and firm risk‑taking behavior

(− 1.53) (− 1.36) (− 5.38) (− 6.76) (1.67) (− 6.88) (− 3.44)

Ln (CEO Tenure) − 0.0011 − 0.0066*** 0.0025*** − 0.0001 − 0.0045*** 0.0053*** 0.0091***
(− 1.48) (− 2.74) (4.40) (− 0.33) (− 6.43) (2.65) (5.66)
Ln (Cash Compensation) 0.0052*** 0.0107*** − 0.0014** 0.0007 0.0030*** − 0.0062*** − 0.0009
(6.32) (4.09) (− 2.26) (1.54) (4.34) (− 2.85) (− 0.50)
Ln (CEO Equity Holding) − 0.0008*** − 0.0019** 0.0006** − 0.0005** 0.0011*** − 0.0037*** 0.0002
(− 2.66) (− 2.12) (2.53) (− 2.55) (3.59) (− 5.02) (0.25)
CEO turnover − 0.0002 − 0.0003 0.0014 − 0.0024** − 0.0086*** − 0.0012 0.0008
(− 0.10) (− 0.06) (1.16) (− 2.43) (− 5.88) (− 0.25) (0.23)
Control variables Yes Yes Yes Yes Yes Yes Yes
Year Fixed Effect Yes Yes Yes Yes Yes Yes Yes
FF48 Fixed Effect Yes Yes Yes Yes Yes Yes Yes
Observations 15,056 15,101 25,095 25,095 23,622 23,261 23,225
Adj R-Square 23.22% 15.78% 19.83% 63.69% 10.51% 22.61% 38.45%

***, **, and * denote the significance levels of 1, 5, and 10%, respectively

13


Table 6  Compare the effects of high (top quartile) and low (both quartile) managerial ability on firm risk-taking behavior while controlling for the effect of managerial incen-
tives
(1) (2) (3) (4) (5) (6) (7)

13
σ(ROA) σ(ROE) CAPEX_TA R&D_TA Acquisition_TA Herfindahl index-ff48 Book Leverage

Hi_MA 0.0054*** 0.0127*** − 0.0025*** 0.0104*** 0.0001 0.0149*** − 0.0020

(4.58) (3.59) (− 3.15) (15.88) (0.10) (5.47) (− 0.88)
Low_MA − 0.0018 − 0.0041 − 0.0036*** − 0.0017*** − 0.0071*** 0.0073** 0.0290***
(− 1.57) (− 1.13) (− 4.61) (− 3.18) (− 7.00) (2.43) (12.82)
Ln (CEO Delta) − 0.0001 − 0.0010 0.0016*** 0.0001 0.0007* − 0.0029*** − 0.0043***
(− 0.21) (− 0.85) (4.99) (0.33) (1.70) (− 2.82) (− 4.88)
Ln (CEO Vega) − 0.0005 0.0001 − 0.0010*** 0.0021*** − 0.0001 0.0028*** 0.0025***
(− 1.50) (0.11) (− 4.94) (14.56) (− 0.50) (3.67) (4.17)
Ln (CEO Age) − 0.0056 − 0.0150 − 0.0150*** − 0.0139*** 0.0055* − 0.0630*** − 0.0261***
(− 1.48) (− 1.32) (− 5.40) (− 6.63) (1.66) (− 6.89) (− 3.43)
Ln (CEO Tenure) − 0.0012 − 0.0067*** 0.0025*** − 0.0002 − 0.0045*** 0.0054*** 0.0093***
(− 1.55) (− 2.80) (4.39) (− 0.52) (− 6.49) (2.67) (5.82)
Ln (CEO Cash Compensation) 0.0052*** 0.0106*** − 0.0013** 0.0007 0.0031*** − 0.0063*** − 0.0013
(6.28) (4.05) (− 2.20) (1.48) (4.43) (− 2.90) (− 0.72)
Ln (CEO Equity Holding) − 0.0008*** − 0.0019** 0.0006** − 0.0005*** 0.0011*** − 0.0037*** 0.0003
(− 2.69) (− 2.13) (2.53) (− 2.62) (3.58) (− 5.04) (0.40)
CEO turnover − 0.0002 − 0.0004 0.0014 − 0.0024** − 0.0087*** − 0.0011 0.0010
(− 0.11) (− 0.08) (1.14) (− 2.50) (− 5.92) (− 0.22) (0.29)
Control variables Yes Yes Yes Yes Yes Yes Yes
Year Fixed Effect Yes Yes Yes Yes Yes Yes Yes
FF48 Fixed Effect Yes Yes Yes Yes Yes Yes Yes
Observations 15,056 15,101 25,095 25,095 23,622 23,261 23,225
Adj R-Square 23.25% 15.81% 19.90% 63.76% 10.61% 22.61% 38.57%

***, **, and * denote the significance levels of 1, 5, and 10%, respectively
K. Yung, C. Chen
Managerial ability and firm risk‑taking behavior

Meckling 1976; Stulz 1990; Fluck and Lynch 1999). On the other hand, a number of stud-
ies on agency conflicts suggest that managers underinvest when investment performance
affects the manager’s reputation (Campbell et  al. 1989; Hirshleifer and Thakor 1992).
Smith and Watts (1992) suggest that managers underinvest because when they face a lim-
ited payout horizon.
To mitigate agency conflicts, corporate governance mechanisms are commonly adopted
by firms to align the interests between shareholders and managers. Among the various gov-
ernance practices, the board of directors plays a pivotal role in the governance of widely
held corporations. Monitoring, advising the top management, and ensuring that the firm is
run in the shareholders’ interest are the fiduciary duties of the board. There is evidence that
the presence of the board of directors has positive impact on firm performance (Yermack
1996; Coles et al. 2008).
In our second robustness test, we control for the effect of corporate governance on man-
agerial risk-taking behavior. Specifically, we add board size to the list of independent vari-
ables in our regression model:
Risk − taking activityit = 𝛽0i + 𝛽1i Managerial Abilityit−1 + 𝛽2i Board Sizeit−1
(5)
+ 𝛽3i Xit + Industry dummies + Year dummies + εit

In Table 7, regression results using Eq. (5) are reported. The results are largely consist-
ent with the findings reported in Table 3. With the exception in column (2), the coefficient
on MA is positive and significant at the 1% level in each regression. The coefficient on
Board Size is negative and significant in columns (1), (5) and (6); suggesting that the board
of directors impose restraints on firm risk-taking. However, consistent with the view of
Wang (2012), we find that the presence of the board of directors is associated with higher
levels of R&D expenditures.
In Table 8, we compare the effects of high managerial ability and low managerial ability
on firm risk-taking while controlling for managerial incentives and board size. The results
show that hi_MA managers are associated with increases of σROE, R&D expenses, and firm
focus. In contrast, low_MA managers are associated with decreases of risk-taking meas-
ured by σROA, CAPEX_TA, R&D_TA, and Acquisitions_TA. The results are largely con-
sistent with the findings reported in earlier tables.

5.3 Managerial ability, firm value, and firm operating performance

In our next investigation, we examine the effect of managerial ability on firm value (meas-
ured by Tobin’s Q). We report the regression results in Tables 9 and 10. As can be seen in
Table 9, the coefficient of managerial ability is positive and significant at the one percent
level in each column. The result persists in column (8) where we control for all the risk-
taking measures and the other independent variables. The economic impact of managerial
ability on firm value is considerable. For example, in column (1), a one standard deviation
in managerial ability is associated with a 13.69% increase in firm value. In column (8), a
one standard deviation in managerial ability is associated with a 10.99% increase in firm
value. The high R-square of each regression suggests that the model specification is accu-
rate and that managerial ability is an important explanatory variable of firm value. Con-
sistent with our result, Demerjian et al. (2012) find that capable CEOs are associated with
improvements in firm performance.
In Table 10, we compare the effects of high managerial ability and low managerial abil-
ity on firm value. As can be seen in the table, the coefficient on hi_MA is positive and

13


13
Table 7  Regressions of managerial ability effect on firm risk-taking behavior while controlling for the effect of corporate governance
(1) (2) (3) (4) (5) (6) (7)
σ(ROA) σ(ROE) CAPEX_TA R&D_TA Acquisition_TA Herfindahl index-ff48 Book Leverage

Managerial ability 0.0176** 0.0179 0.0078* 0.0377*** 0.0231*** 0.0730*** − 0.1083***

(1.99) (0.68) (1.88) (10.53) (4.26) (6.14) (− 8.86)
Board size − 0.0008* 0.0005 − 0.0001 0.0009*** − 0.0006* − 0.0030*** − 0.0005
(− 1.81) (0.36) (− 0.45) (5.04) (− 1.88) (− 3.42) (− 0.73)
Control variables Yes Yes Yes Yes Yes Yes Yes
Year Fixed Effect Yes Yes Yes Yes Yes Yes Yes
FF48 Fixed Effect Yes Yes Yes Yes Yes Yes Yes
Observations 4959 4989 11,154 11,154 10,616 10,454 10,413
Adj R-Square 17.69% 12.71% 19.14% 60.84% 9.90% 20.92% 37.93%

***, **, and * denote the significance levels of 1, 5, and 10%, respectively
K. Yung, C. Chen
 Table 8  Compare the effects of high (top quartile) and low (bottom quartile) managerial ability on firm risk-taking behavior while controlling for the effect of corporate governance
(1) (2) (3) (4) (5) (6) (7)
σ(ROA) σ(ROE) CAPEX_TA R&D_TA Acquisition/AT Herfindahl index Book Leverage

Hi_MA 0.0035 0.0148* 0.0004 0.0082*** 0.0018 0.0215*** − 0.0045

(1.29) (1.81) (0.36) (7.77) (1.08) (5.36) (− 1.17)
Low_MA − 0.0049** 0.0066 − 0.0053*** − 0.0032*** − 0.0045*** 0.0054 0.0244***
(− 2.02) (0.90) (− 5.28) (− 3.83) (− 2.91) (1.19) (7.07)
Ln (CEO Delta) − 0.00109 − 0.0024 0.0028*** − 0.0004 0.0019*** − 0.0058*** − 0.0005
(− 1.07) (− 0.79) (5.87) (− 1.05) (2.92) (− 3.32) (− 0.35)
Ln (CEO Vega) 0.0006 0.0010 − 0.0014*** 0.0022*** 0.0002 0.0062*** − 0.0006
(0.84) (0.48) (− 4.98) (10.09) (0.47) (5.57) (− 0.68)
Ln (CEO Age) − 0.00326 − 0.0335 − 0.0023 − 0.0183*** 0.0026 − 0.0531*** − 0.0378***
(− 0.37) (− 1.32) (− 0.59) (− 5.56) (0.45) (− 3.68) (− 3.04)
Managerial ability and firm risk‑taking behavior

Ln (CEO Tenure) − 0.00408** − 0.0138** 0.0007 0.0001 − 0.0064*** 0.0025 0.0150***

(− 2.12) (− 2.37) (0.78) (0.11) (− 5.18) (0.78) (5.37)
Ln (CEO Cash Compensation) 0.0000** 0.0000 0.0000 0.0000 0.0000 0.0000*** 0.0000*
(2.15) (1.61) (− 0.85) (− 0.48) (− 0.33) (− 6.28) (− 1.85)
Ln (CEO Equity Holding) 0.0000 0.0000 0.0000** 0.0000*** 0.0000 0.0000** 0.0000***
(1.00) (− 0.08) (− 2.23) (4.86) (0.41) (− 2.40) (− 7.91)
CEO turnover − 0.0047 − 0.0146 − 0.0005 − 0.0022 − 0.0145*** 0.0030 0.0059
(− 1.16) (− 1.19) (− 0.32) (− 1.46) (− 6.12) (0.43) (1.02)
Board size − 0.0012** − 0.0014 0.0003 0.0007*** − 0.0008** − 0.0040*** − 0.0009
(− 2.47) (− 0.97) (1.44) (3.59) (− 2.51) (− 4.32) (− 1.24)
Control variables Yes Yes Yes Yes Yes Yes Yes
Year Fixed Effect Yes Yes Yes Yes Yes Yes Yes
FF48 Fixed Effect Yes Yes Yes Yes Yes Yes Yes
Observations 4503 4536 10,377 10,377 9868 9733 9689
Adj R-Square 18.83% 13.70% 18.62% 61.52% 10.35% 22.63% 38.70%

***, **, and * denote the significance levels of 1, 5, and 10%, respectively

13
 K. Yung, C. Chen

significant at the one percent level in all the eight columns; the coefficient on low_MA is
negative and significant at the one percent level in the first seven columns. The results sug-
gest that high-ability managers are more likely to improve firm value whereas low-ability
managers are more likely to destroy firm value.1 A plausible reason is that low-ability man-
agers are prone to make mistakes in investment decisions. Another possible explanation is
that low-ability managers are more likely to emphasize private motives, resulting in firm-
value reducing decisions. We repeat the analysis in Tables  9 and 10 and control for the
effects of managerial incentives and board size; we find similar results and the results are
therefore not tabulated.
In Table 11, we provide information on the effect of managerial ability on firm operat-
ing performance. We divide the sample period into three subperiods surrounding the finan-
cial crisis of 2007–2009: (1) pre-financial crisis; (2) during financial crisis; and (3) post
financial crisis. We compare the operating performance of firms with high MA scores and
firms with low MA scores. The performance measures include return on assets (ROA),
return on equity (ROE), sales volume, sales growth, and earnings (EBIT_TA). We also
provide information on dividend payout and leverage.
Panel A of Table 11 shows that during the pre-crisis period, hi_MA firms have signifi-
cantly higher ROA, ROE, and earnings than low_MA firms. Hi_MA and low_MA firms
have comparable dividend payout ratios, but hi_MA firms have a much lower leverage.
The results remain consistent during the financial crisis of 2007–2009. During this period,
both hi_MA and low_MA firms suffer and have negative ROA and ROE, but hi_MA firms
suffer considerably less than low_MA firms. In the post-crisis period, ROA and ROE turn
positive for hi_MA firms but remain negative for low_MA firms. During the post-crisis
period, the average earnings of hi_MA firms is almost three times the average earnings of
low_MA firms. As in the other subperiods, hi_MA firms continue to have a lower leverage
than low_MA firms. In sum, hi_MA firms are in general more profitable, have higher earn-
ings, and have lower leverage ratios.

6 Summary and conclusion

Conventional models ignore the role of managerial heterogeneity in firm decisions as man-
agers are expected to act rationally and follow the mandate of firm value maximization. We
examine the relation between managerial ability and the risk choices in corporate invest-
ment. Our view is that in addition to the effects of previously examined determinants, the
risk choices are affected not just by the managers’ explicit mandate to maximize firm value,
but also by the ability of the manager in managing the firm. We add to the literature by
showing that managerial ability significantly affects firm policies. In contrast to previous
studies that find a positive linear relationship between managerial ability and firm invest-
ment behavior, we find that high-ability managers and low-ability managers have opposite
effects on firm risk-taking behavior and firm value. High-ability managers are in general
receptive to risk-taking whereas low-ability managers refrain from risk-taking. High-
ability managers cut capital expenditures but spends significantly more on research and

1
  We thank an anonymous referee for pointing out that managerial ability may also be a measure of mana-
gerial risk aversion. Given that our results show a positive relation between hi_MA and firm value, we think
that managerial ability is less likely a measure of risk aversion because firm value may not be enhanced by
aggressive managers.

13
 Table 9  Regressions of managerial ability effect on firm value
(1) (2) (3) (4) (5) (6) (7) (8)

Managerial Ability 2.4511*** 2.6400*** 2.0430*** 2.2273*** 2.1947*** 0.9672*** 2.7304*** 1.9671***
(28.58) (31.34) (35.86) (41.88) (39.95) (4.81) (34.04) (7.23)
σ(ROA) 2.2233*** 1.4166***
(21.02) (8.52)
σ(ROE) 0.4485*** 0.1681***
(13.10) (3.07)
CAPEX_TA 1.6053*** 1.3979***
(19.76) (13.57)
Managerial ability and firm risk‑taking behavior

R&D_TA 3.3818*** 3.5140***

(31.51) (20.20)
Acquisition_TA − 1.7872*** − 0.9122***
(− 28.57) (− 10.07)
Herfindahl index-ff48 0.2466*** 0.0952***
(11.58) (3.62)
Book Leverage − 1.2412*** − 1.0966***
(− 53.50) (− 31.37)
Control variables Yes Yes Yes Yes Yes Yes Yes Yes
Year Fixed Effect Yes Yes Yes Yes Yes Yes Yes Yes
FF48 Fixed Effect Yes Yes Yes Yes Yes Yes Yes Yes
Observations 51,889 51,978 90,592 90,592 87,259 23,291 25,095 12,390
Adj R-Square 32.32% 32.24% 31.90% 33.11% 31.86% 47.49% 48.29% 53.71%

***, **, and * denote the significance levels of 1, 5, and 10%, respectively

13


Table 10  Compare the effects of high (top quartile) and low (bottom quartile) managerial ability on firm value
(1) (2) (3) (4) (5) (6) (7) (8)

13
Hi_MA 0.4024*** 0.4290*** 0.4167*** 0.3570*** 0.4107*** 0.4148*** 0.3775*** 0.3263***
(28.07) (29.50) (36.59) (31.12) (35.50) (35.31) (33.60) (21.70)
Low_MA − 0.0734*** − 0.0701*** − 0.0685*** − 0.0777*** − 0.0809*** − 0.0796*** − 0.0315*** − 0.0204
(− 5.72) (− 5.34) (− 6.78) (− 7.82) (− 7.93) (− 7.58) (− 3.14) (− 1.53)
σ(ROA) 2.2399*** 1.3836***
(20.94) (8.31)
σ(ROE) 0.4460*** 0.1850***
(12.92) (3.44)
CAPEX_TA 1.6419*** 1.4119***
(20.51) (13.88)
R&D_TA 3.3532*** 3.4702***
(31.41) (20.19)
Acquisition_TA − 1.7757*** − 0.9177***
(− 28.19) (− 9.95)
Herfindahl index-ff48 0.2470*** 0.1072***
(11.76) (4.12)
Book Leverage − 1.22431*** − 1.1002***
(− 53.64) (− 32.28)
Control variables Yes Yes Yes Yes Yes Yes Yes Yes
Year Fixed Effect Yes Yes Yes Yes Yes Yes Yes Yes
FF48 Fixed Effect Yes Yes Yes Yes Yes Yes Yes Yes
Observations 51,889 51,978 90,592 90,592 87,259 85,647 90,462 44,829
Adj R-Square 31.61% 31.46% 31.32% 32.49% 31.31% 31.17% 32.84% 35.74%

***, **, and * denote the significance levels of 1, 5, and 10%, respectively
K. Yung, C. Chen
Managerial ability and firm risk‑taking behavior

Table 11  Managerial ability and firm operating performance

Hi_MA Low_MA High-Low Difference Difference between
between means medians (p value)
(t-value)

Panel A: pre-crisis (Before 2007)

 ROA 0.0014 − 0.0627 0.0641 34.15 <.0001
 ROE − 0.0721 − 0.2442 0.1721 23.67 <.0001
 Log(Sales) 4.8789 4.8528 0.0261 1.28 <.0001
 Sales growth 0.1214 0.1346 − 0.0132 − 4.37 <.0001
 EBIT_TA 0.0957 0.0343 0.0614 27.39 <.0001
 Dividends/earnings 0.0612 0.0697 − 0.0085 − 0.84 0.8563
 Leverage 0.1820 0.2834 − 0.1014 − 59.74 <.0001
Panel B: during-crisis (from 2007 to 2009)
 ROA − 0.0050 − 0.0613 0.0563 8.92 <.0001
 ROE − 0.0954 − 0.2414 0.1460 5.9 <.0001
 Log(Sales) 5.6496 6.2693 − 0.6197 − 9.66 <.0001
 Sales growth 0.0636 0.0424 0.0212 2.21 <.0001
 EBIT_TA 0.0902 0.0304 0.0598 8.54 <.0001
 Dividends/earnings 0.0812 − 0.0357 0.1169 1.09 0.0009
 Leverage 0.1409 0.2608 − 0.1198 − 21.99 <.0001
Panel C: post-crisis (After 2009)
 ROA 0.0288 − 0.0443 0.0731 15.72 <.0001
 ROE 0.0024 − 0.163 0.1658 9.39 <.0001
 Log(Sales) 6.4825 6.0609 0.4216 7.6 0.0811
 Sales growth 0.0993 0.1119 − 0.0126 − 1.75 0.0093
 EBIT_TA 0.1171 0.0398 0.0772 15.18 <.0001
 Dividends/earnings 0.1478 0.0713 0.0764 1.5 0.1097
 Leverage 0.1700 0.2561 − 0.0860 − 18.33 <.0001

development projects; low-ability managers reduce both capital expenditures and research
and development expenses significantly. High-ability managers are associated with higher
levels of firm focus than low-ability managers. Our results are robust after taking into con-
sideration the effects of managerial incentives and corporate governance on CEO behavior.
Our results also show that high-ability managers are associated with increases in firm value
whereas low-ability managers are associated with decreases in firm value. The observation
suggests that managerial ability is integral to the long-term success of a firm. The result
also provides justification for the lucrative compensation packages offered to attract capa-
ble managers.

References
Amihud Y, Lev B (1981) Risk reduction as a managerial motive for conglomerate mergers. Bell J Econ
12:605–617
Andreou PC, Philip D, Robejsek P (2016) Bank liquidity creation and risk-taking: does managerial ability
matter? J Bus Finance Account 43:226–259

13
 K. Yung, C. Chen

Baker M, Wurgler J (2012) Behavioral corporate finance: a current survey. In: Eckbo E (ed) Handbook of
the economics of finance, vol 2. Elsevier, New York
Bamber LS, Jiang J, Wang IY (2010) What’s my style? The influence of top managers on voluntary corpo-
rate financial disclosure. Account Rev 85:1131–1162
Bandura A (1997) Self-efficacy: the exercise of control. Macmillan, New York
Bargeron LL, Lehn KM, Zutter CJ (2010) Sarbanes-Oxley and corporate risk-taking. J Account Econ
49:34–52
Bates TW, Kahle KM, Stulz RM (2009) Why do US firms hold so much more cash than they used to? J
Finance 64:1985–2021
Berger PG, Ofek E (1995) Diversification’s effect on firm value. J Financ Econ 37:39–65
Berk JB, Stanton R (2007) Managerial ability, compensation, and the closed-end fund discount. J Finance
62:529–556
Bertrand M, Mullainathan S (2003) Enjoying the quiet life? Corporate governance and managerial prefer-
ences. J Polit Econ 111:1043–1075
Bertrand M, Schoar A (2003) Managing with style: the effect of managers on firm policies. Q J Econ
118:1169–1208
Bhagat S, Welch I (1995) Corporate research & development investments international comparisons. J
Account Econ 19:443–470
Bhagat S, Bolton B, Subramanian A (2012) Manager characteristics and capital structure: theory and evi-
dence. J Financ Quant Anal 46:1581–1627
Bloom N, Van Reenen J (2009) Bossonomics: the economics of management and productivity. Centre for
Economic Performance, LSE
Boholm A (1998) Comparative studies of risk perception: a review of twenty years of research. J Risk Res
1:135–163
Brick IE, Palmon O, Wald JK (2012) Too much pay-performance sensitivity? Rev Econ Stat 94:287–303
Bromiley P (1991) Testing a causal model of corporate risk taking and performance. Acad Manag J
34:37–59
Camerer C (2003) Behavioral game theory: experiments in strategic interaction. Princeton University Press,
Princeton
Camerer CF, Ho T-H, Chong J-K (2004) A cognitive hierarchy model of games. Q J Econ 119:861–898
Campbell TS, Chan Y-S, Marino AM (1989) Incentive contracts for managers who discover and manage
investment projects. J Econ Behav Organ 12:353–364
Carpenter JN (2000) Does option compensation increase managerial risk appetite? J Finance 55:2311–2331
Charnes A, Cooper WW, Rhodes E (1978) Measuring the efficicency of decision making units. Eur J Oper
Res 2:429–444
Chemmanur TJ, Paeglis I, Simonyan K (2009) Management quality, financial and investment policies, and
asymmetric information. J Financ Quant Anal 44:1045–1079
Chen Y, Podolski EJ, Veeraraghavan M (2015) Does managerial ability facilitate corporate innovative suc-
cess? J Empir Finance 34:313–326
Choi W, Han S, Jung SH, Kang T (2015) CEO’s operating ability and the association between accruals and
future cash flows. J Bus Finance Account 42:619–634
Coles JL, Daniel ND, Naveen L (2006) Managerial incentives and risk-taking. J Financ Econ 79:431–468
Coles JL, Lemmon ML, Wang YA (2008) The joint determinants of managerial ownership, board independ-
ence, and firm performance. SSRN Working Paper
Comment R, Jarrell GA (1995) Corporate focus and stock returns. J Financ Econ 37:67–87
Core J, Guay W (2002) Estimating the value of employee stock option portfolios and their sensitivities to
price and volatility. J Account Res 40:613–630
Costa-Gomes MA, Crawford VP (2006) Cognition and behavior in two-person guessing games: an experi-
mental study. Am Econ Rev 96:1737–1768
Cremers KM, Petajisto A (2009) How active is your fund manager? A new measure that predicts perfor-
mance. Rev Financ Stud 22:3329–3365
Culver J, Burke W, Yasui Y, Durfy S, Press N (2001) Participation in breast cancer genetic counseling: the
influence of educational level, ethnic background, and risk perception. J Genet Couns 10:215–231
Custódio C, Ferreira MA, Matos P (2013) Generalists versus specialists: lifetime work experience and chief
executive officer pay. J Financ Econ 108:471–492
Demerjian PR, Lev B, McVay S (2012) Quantifying managerial ability: a new measure and validity tests.
Manag Sci 58:1229–1248
Demerjian PR, Lev B, Lewis MF, McVay SE (2013) Managerial ability and earnings quality. Account Rev
88:463–498

13
Managerial ability and firm risk‑taking behavior

Faccio M, Marchica M-T, Mura R (2011) Large shareholder diversification and corporate risk-taking. Rev
Financ Stud 24:3601–3641
Fang VW, Tian X, Tice S (2014) Does stock liquidity enhance or impede firm innovation? J Finance
69:2085–2125
Fluck Z, Lynch AW (1999) Why do firms merge and then divest? A theory of financial synergy. J Bus
72:319–346
Goldfarb A, Xiao M (2011) Who thinks about the competition? Managerial ability and strategic entry in US
local telephone markets. Am Econ Rev 101:3130–3161
Gormley TA, Matsa DA (2016) Playing it safe? Managerial preferences, risk, and agency conflicts. J Financ
Econ 122:431–455
Graham JR, Li S, Qiu J (2012) Managerial attributes and executive compensation. Rev Financ Stud
25:144–186
Guay WR (1999) The sensitivity of CEO wealth to equity risk: an analysis of the magnitude and determi-
nants. J Financ Econ 53:43–71
Guo L, Jalal A, Khaksari S (2015) Bank executive compensation structure, risk taking and the financial cri-
sis. Rev Quant Finance Account 45:609–639
Halek M, Eisenhauer JG (2001) Demography of risk aversion. J Risk Insur 68:1–24
Hambrick DC (2007) Upper echelons theory: an update. Acad Manag Rev 32:334–343
Hambrick DC, Mason PA (1984) Upper echelons: the organization as a reflection of its top managers. Acad
Manag Rev 9:193–206
Haugen RA, Senbet LW (1981) Resolving the agency problems of external capital through options. J
Finance 36:629–647
Hirshleifer D, Thakor AV (1992) Managerial conservatism, project choice, and debt. Rev Financ Stud
5:437–470
Jensen MC, Meckling WH (1976) Theory of the firm: managerial behavior, agency costs and ownership
structure. J Financ Econ 3:305–360
Jenter D, Lewellen K (2015) CEO preferences and acquisitions. J Finance 70:2813–2852
John K, Litov L, Yeung B (2008) Corporate governance and risk-taking. J Finance 63:1679–1728
Kaplan SN, Klebanov MM, Sorensen M (2012) Which CEO characteristics and abilities matter? J Finance
67:973–1007
Koijen RSJ (2014) The cross-section of managerial ability, incentives, and risk preferences. J Finance
69:1051–1098
Lambert RA, Larcker DF, Verrecchia RE (1991) Portfolio considerations in valuing executive compensa-
tion. J Account Res 29:129–149
Low A (2009) Managerial risk-taking behavior and equity-based compensation. J Financ Econ 92:470–490
Maccrimmon K, Wehrung DA (1986) The management of uncertainty: taking risks. Free Press, New York
Malmendier U, Tate G (2005) CEO overconfidence and corporate investment. J Finance 60:2661–2700
Malmendier U, Tate G (2008) Who makes acquisitions? CEO overconfidence and the market’s reaction. J
Financ Econ 89:20–43
March JG, Shapira Z (1987) Managerial perspectives on risk and risk taking. Manag Sci 33:1404–1418
Mishra DR (2014) The dark side of CEO ability: CEO general managerial skills and cost of equity capital. J
Corp Finance 29:390–409
Murphy KJ, Zabojnik J (2007) Managerial capital and the market for CEOs. SSRN Working Paper
Nuthall P (2001) Managerial ability—a review of its basis and potential improvement using psychological
concepts. Agric Econ 24:247–262
Nuthall P (2009) Modelling the origins of managerial ability in agricultural production. Aust J Agric Resour
Econ 53:413–436
Porter ME (1987) From competitive advantage to corporate strategy. Harv Bus Rev 65:43–59
Ross SA (2004) Compensation, incentives, and the duality of risk aversion and riskiness. J Finance
59:207–225
Shapira Z (1986) Risk in managerial decision making. Hebrew University working paper
Sjöberg L, Drottz-Sjöberg BM (1991) Knowledge and risk perception among nuclear power plant employ-
ees. Risk Anal 11:607–618
Smith CW, Watts RL (1992) The investment opportunity set and corporate financing, dividend, and com-
pensation policies. J Financ Econ 32:263–292
Stulz R (1990) Managerial discretion and optimal financing policies. J Financ Econ 26:3–27
Tian X, Wang TY (2014) Tolerance for failure and corporate innovation. Rev Financ Stud 27:211–255
Wang C-J (2012) Board size and firm risk-taking. Rev Quant Finance Account 38:519–542

13
 K. Yung, C. Chen

Wang Y, Zhou W, Chang K-C (2013) Effect of decision makers’ education level on their corporate risk tak-
ing. Soc Behav Personal Int J 41:1225–1229
Yermack D (1996) Higher market valuation of companies with a small board of directors. J Financ Econ
40:185–211
Yim S (2013) The acquisitivenss of youth: CEO age and acquisition behavior. J Financ Econ 108:250–273

13